Polar Code List Decoding with Neural Likelihood Fusion
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing data storage devices face challenges in efficiently decoding data due to errors or corruption, particularly when the number of errors exceeds the error correction capabilities of current ECC mechanisms.
Innovation Solution
The integration of a neural network with a decoder, such as a polar code decoder, to analyze previously decoded data and provide likelihood values that enhance the decoder's ability to correct errors and improve decoding robustness.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If traditional ECC mechanisms are used for error correction, then the decoding process is simple and fast, but the error correction capability is insufficient when errors exceed certain thresholds
Solution Approach 1:
A neural network is introduced as an intermediary component between the received codeword and the polar code decoder. The neural network processes the input codeword and generates enhanced likelihood values that are fed into the decoder, thereby improving error correction capability without requiring the decoder itself to become more complex
Solution Approach 2:
The decoding system is segmented into two distinct functional components: a neural network module for generating likelihood values and a polar code decoder module for performing the actual decoding. This segmentation allows each component to be optimized independently, with the neural network handling the complex pattern recognition while the decoder maintains its established simple and fast operation
2Reliability
If neural network integration is added to enhance error correction, then the decoding robustness improves, but the computational complexity and processing time increase
Solution Approach 1:
The neural network performs preliminary processing of the received codeword before it reaches the polar code decoder. By pre-computing the likelihood values using the neural network, the actual decoding process can proceed more efficiently with higher robustness, as the difficult pattern recognition work is already done in advance
Solution Approach 2:
The system changes the parameter representation of the input data by transforming the raw codeword into neural network-generated likelihood values. This parameter transformation enables the decoder to work with pre-processed, more informative data that improves robustness while maintaining efficient decoding operations
Data Source
Figure 1
Figure 2
Figure 3
AI summary
Systems and methods are disclosed for decoding data. A first block of data may be obtained from a storage medium or received from a computing device. The first block of data includes a first codeword generated based on an error correction code. A first set of likelihood values is obtained from a neural network. The first set of likelihood values indicates probabilities that the first codeword will be decoded into one of a plurality of decoded values. A second set of likelihood values is obtained from a decoder based on the first block of data. The second set of likelihood values indicates probabilities that the first codeword will be decoded into one of the plurality of decoded values. The first codeword is decoded to obtain a decoded value based on the first set of likelihood values and the second set of likelihood values.